Performance of Multiple Pretrained BERT Models to Automate and Accelerate Data Annotation for Large Datasets

PurposeTo develop and evaluate domain-specific and pretrained bidirectional encoder representations from transformers (BERT) models in a transfer learning task on varying training dataset sizes to annotate a larger overall dataset. Materials and MethodsThe authors retrospectively reviewed 69 095 ano...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Radiology. Artificial intelligence 2022-07, Vol.4 (4), p.e220007-e220007
Hauptverfasser:	Tejani, Ali S., Ng, Yee S., Xi, Yin, Fielding, Julia R., Browning, Travis G., Rayan, Jesse C.
Format:	Artikel
Sprache:	eng
Schlagworte:	Original Research
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!